Artificial Intelligence for Quantitative MRI in Prostate Cancer: From Detection to Precision Care

Prostate cancer imaging is increasingly moving beyond visual interpretation toward quantitative, AI-supported decision-making. Multiparametric and biparametric MRI have become central to lesion detection, biopsy targeting, local staging, and risk assessment, while whole-body diffusion MRI supports the evaluation of metastatic disease and treatment response. Radiomics, apparent diffusion coefficient measurements, machine learning, and deep learning are further expanding the potential of MRI to provide objective biomarkers and individualized predictions. However, variation in imaging protocols, patient populations, annotation practices, and technical pipelines continues to limit the reproducibility and generalizability of AI models. Questions also remain regarding the clinical relevance of quantitative metrics, the appropriate design of external validation studies, and the integration of AI tools into existing radiology workflows. Current research is addressing these challenges through improved model development, multimodal approaches, harmonization strategies, and clinically oriented evaluation frameworks, but further investigation is needed to establish robust, transparent, and clinically useful solutions.This Research Topic aims to advance understanding of the computational methods underpinning AI-supported quantitative MRI in prostate cancer. It will examine how machine learning, deep learning, radiomics, and quantitative imaging biomarkers can improve detection, risk stratification, biopsy guidance, staging, treatment monitoring, and outcome prediction. Emphasis will be placed on methodological rigor, reproducibility, external validation, interpretability, and clinical utility. The Research Topic will also explore how AI tools can be integrated into clinical workflows and whether they provide meaningful benefits beyond established imaging and clinical approaches.To gather further insights into AI and computational methods applied to quantitative prostate cancer MRI, we welcome articles addressing, but not limited to, the following themes:• Machine learning and deep learning models for detecting clinically significant prostate cancer• AI-assisted lesion characterization, risk stratification, and biopsy targeting• Radiomics and quantitative imaging biomarkers, including ADC-based metrics• Automated extraction and analysis of quantitative MRI features• Deep learning approaches for metastatic assessment using whole-body diffusion MRI• AI-supported local staging and assessment of treatment response• Automated or AI-assisted structured reporting and clinical decision support• Multimodal AI models integrating quantitative MRI features with clinical, pathological, or genomic data• Reproducibility, robustness, explainability, and fairness of AI models• Internal, external, multicenter, and prospective validation studies• Data harmonization and methods for addressing scanner and protocol variability• Clinical workflow integration, implementation, and evaluation of AI tools• Comparisons between AI-based methods and established radiological or clinical approaches• Study designs, reporting standards, and regulatory considerations for clinical translationWe welcome Original Research, Systematic Reviews, Methods papers, Study Protocols, and Perspectives, particularly those addressing the validation, reproducibility, interpretability, or workflow integration of AI tools in this field. Topic editors: Lei Hu, Jiule Ding Source: https://www.frontiersin.org/research-topics/84205/artificial-intelligence-for-quantitative-mri-in-prostate-cancer-from-detection-to-precision-care